CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Conclusions: This is the first report, to the best of our knowledge, of a high-throughput complete HIV genome sequencing and analysis pipeline in Africa. The genetic diversity of HIV variants in this population is high and is mediated primarily by multiple introductions of HIV. Interventions therefore must be cognizant of the dynamics that drive these independent introductions in order to impact on going HIV transmission. 257 Pan-HIV Next-Gen Sequencing Strategy for Viral Surveillance Michael G. Berg 1 ; JulieYamaguchi 1 ; Elodie Alessandri-Gradt 2 ; Jean-Christophe Plantier 2 ; Catherine Brennan 1 1 Abbott Laboratories, Abbott Park, IL, US; 2 Virology Unit, National Reference for HIV, Rouen, France Background: The complexity of HIV strains has increased significantly due to natural evolution and inter-subtype recombination; recombination is now a worldwide problem. Thus, complete genome sequencing is essential to monitor HIV diversity accurately within populations. Next-generation sequencing (NGS) has the potential to revolutionize strategies for HIV surveillance. We have developed a universal method that permits full genome sequencing of all HIV-1 groups (M, N, O, and P) and HIV-2. Methods: Reverse transcription primers, designed in conserved regions of HIV and spaced at 1.5-2kb intervals, fuse viral sequences to a common adaptor (SMART) sequence. This same adaptor is added to the 3’ end of the cDNA to permit PCR amplification of libraries, which are then tagmented with Nextera XT for multiplexing and sequencing on an Illumina MiSeq. HIV sequences are extracted and assembled in CLC-Bio software (Qiagen) and classified by phylogenetic analysis using PHYLIP and SIMPLOT. Results: Broad application of the approach was demonstrated using a panel of virus isolates (n=47) derived from cell culture that included 27 group M (different subtypes and CRFs), 16 O, 2 N, 1 P, and 1 HIV-2. In a single run multiplexing 23 libraries, 100% genome coverage was obtained for each at a median depth of 2100X, with HIV reads comprising 9.4% (median) of the total. A Cameroonian HIV-1 non-subtype B specimen was used to optimize the protocol for plasma. An NGS run of 8 high titer (>5.0 log copies/ml) clinical specimens, infected with diverse group M subtypes, yielded 96-100% coverage for each at a median 433X depth. Method sensitivity, demonstrated by serial dilution, showed that >50% of a genome could be obtained from a clinical specimen with a viral load >3.8 log copies/ml (100 RNA copies input). Due to inherent variability in sample background, coverage varied widely among specimens with viral loads <4.5 log . Nevertheless, sufficient sequence was obtained for strain classification. From the 55 novel full-length HIV sequences determined in this study, 5 are unique recombinants. Conclusions: The HIV-SMART approach harnesses the specificity of HIV-directed priming without a prior i knowledge of the viral strain present. This technology provides an unparalleled opportunity to identify diverse HIV strains in patient specimens and to determine phylogenetic classification based on the entire viral genome, illustrating the utility of NGS for viral surveillance. 258 PCR-Free Full Genome Characterization of Diverse HIV-1 Strains by Nextgen Sequencing Viswanath Ragupathy 1 ; Feng Gao 2 ; Ana Sanchez 2 ; Marco Schito 3 ;Thomas Denny 2 ; Michael Busch 4 ; Jiangqin Zhao 1 ; Christelle Mbondji 1 ; SaiVikramVemula 1 ; Indira Hewlett 1 1 US Food and Drug Administration, Silver Spring, MD, US; 2 Duke Human Vaccine Institute and Departments of Medicine, Duke University Medical Center, Durham, NC, US; 3 Henry Jackson Foundation, DAIDS, NIAID, Bethesda, MD, US; 4 Blood Systems Research Institute/University of California San Francisco, San Francisco, CA, US Background: HIV-1 genotyping is an important tool for clinical and epidemiological studies. High level of genetic variation, recombination and mutations pose difficulty in successful PCR amplification of HIV-1 genomes. In addition, new emerging subtypes may not be detected with standard PCR primers. Here, we report a novel PCR-free multiplex method for characterization of full length HIV-1 genomes (~9.7kb) using the Nextgen RNA Seq approach. Methods: A total of 27 diverse HIV-1 strains representing subtypes A-G, CRFs, URFs and Group O were obtained from the Global HIV-1 diversity panel that was assembled at the Duke EQAPOL. Viral RNA was extracted, reverse transcribed as described in the Illumina Truseq RNA Kit and sequenced using the MiSeq platform. Sequence reads were quality filtered and reference mapped using CLC genomic work bench software v6.0.4. Consensus sequences were generated for each virus and used for phylogenetic tree analysis using the neighbor-joining method based on the Kimura two-parameter substitution model and recombination patterns were determined using Simplot. Drug resistance was inferred from the Stanford HIV drug resistance program, and co-receptor usage was determined using the Geno2Pheno (g2p) 5-10% FPR. Results: The multiplex RNA sequencing approach yielded >10000x coverage for each of the viral genomes. Pools of viral isolates were de-multiplexed and discriminated using bio-informatics. After filtering reads specific for HIV-1, each position in the viral genome had >1000x coverage. This approach enabled reconstruction of whole genome HIV-1 haplotypes accurately including flanking LTRs. Analysis of full HIV-1 genome sequences using Simplot correctly identified 15 pure subtypes, one Group O virus, and recombination patterns of 8 CRFs and 3 URFs. All these HIV subtypes identified were comparable to Sanger sequencing. In addition, this approach revealed NNRTI, integrase and protease drug- specific minor variants and drug resistance mutations with >1000x coverage. The g2p analysis predicted 89% of isolates as being R5 tropic, and the remaining were identified as X4 tropic. Conclusions: We have developed a reliable, PCR-free and multiplexing approach to characterize whole HIV-1 genome sequences. This novel PCR-free method can be used for characterization of new, emerging unknown subtypes or recombinants and to reduce PCR-derived sequence errors. The multiplexing approach makes this NGS method more cost- effective and less labor-intensive than conventional methods. 259 Full-Length Env Deep Sequencing in a Donor With Broadly Neutralizing V1/V2 Antibodies Ben Murrell 1 ; Melissa Laird 2 ; Elise Landais 3 ; Caroline Ignacio 1 ; Ellen Paxinos 2 ; Pham Phung 4 ; Sergei L. Kosakovsky Pond 1 ; Douglas D. Richman 1 ; Pascal Poignard 3 ; Davey M. Smith 1 1 University of California San Diego, La Jolla, CA, US; 2 Pacific Biosciences, Menlo Park, CA, US; 3 The International AIDS Vaccine Initiative Neutralizing Antibody Center, La Jolla, CA, US; 4 LabCorp, South San Francisco, CA, US Background: Understanding the co-evolution of HIV populations and broadly neutralizing antibody (bNAb) lineages may inform vaccine design. Novel long-read, next- generation sequencing methods allow, for the first time, full-length deep sequencing of HIV env populations. Methods: We longitudinally examined env populations (12 time points) in a subtype A infected individual from the IAVI primary infection cohort (Protocol C) who developed bNAbs (62% ID50>50 on a diverse panel of 105 viruses) targeting the V1/V2 region. We developed a Pacific Biosciences single molecule, real-time sequencing protocol to deeply sequence full-length env from HIV RNA. Bioinformatics tools were developed to align env sequences, infer phylogenies, and interrogate escape dynamics of key residues and glycosylation sites. PacBio env sequences were compared to env sequences generated through amplification and cloning. Env dynamics were interpreted in the context of the development of a V1/V2-targeting bNAb lineage isolated from the donor. Results: We collected a median of 6799 high quality full-length env sequences per timepoint (median per-base accuracy of 99.7%). A phylogeny inferred with PacBio and 100 cloned env sequences (10 time points) found cloned env sequences evenly distributed among PacBio sequences. Phylogenetic analyses also revealed a potential transient intra-clade superinfection visible as a minority variant (~5%) at 9 months post-infection (MPI), and peaking in prevalence at 12MPI (~64%), just preceding the development of heterologous neutralization. Viral escape from the bNAb lineage was evident at V2 positions 160, 166, 167, 169 and 181 (HxB2 numbering), exhibiting several distinct escape pathways by 40MPI.

Poster Abstracts

226

CROI 2015